This study proposes a novel approach in response to the persistent challenge of achieving precise autofocus in Digital Lensless Holographic Microscopy (DLHM). It involves employing an enhanced Bluestein algorithm to simulate DLHM holograms under a variety of conditions, spanning amplitude-only, phase-only, and amplitude-phase objects. These simulated holograms are used to assess the performance of autofocus metrics, including the Dubois and Spectral Dubois metrics, gradient and variance-based approaches, and lastly a learning-based model. By considering the variety of sample types and geometrical configurations, this study delves into the robustness and limitations of these metrics across diverse scenarios. This research reveals different performances depending on sample characteristics, offering valuable insights into selecting the most suitable autofocus metric, which is a demanding step in practical DLHM applications.
This study proposes a novel approach in response to the persistent challenge of achieving precise autofocus in Digital Lensless Holographic Microscopy (DLHM). It involves employing an enhanced Bluestein algorithm to simulate DLHM holograms under a variety of conditions, spanning amplitude-only, phase-only, and amplitude-phase objects. These simulated holograms are used to assess the performance of autofocus metrics, including the Dubois and Spectral Dubois metrics, gradient and variance-based approaches, and lastly a learning-based model. By considering the variety of sample types and geometrical configurations, this study delves into the robustness and limitations of these metrics across diverse scenarios. This research reveals different performances depending on sample characteristics, offering valuable insights into selecting the most suitable autofocus metric, which is a demanding step in practical DLHM applications.
This study introduces an innovative approach to volumetric polarimetry, introducing a method that combines Gabor inline holography with conventional polarimetric setups to compute the complete Mueller Matrix of a sample. The study includes the validation of the technique through calibration targets and extends to complex volumetric samples, showcasing the potential of this method for characterizing intricate polarization properties in three-dimensional specimens.
Owing to its high resolution, sensitivity, imaged field of view, and frame rate acquisition, Digital Holographic Microscopy (DHM) stands out among the Quantitative phase imaging (QPI) techniques to reconstruct high-resolution phase images from micrometer-sized samples, providing information about the sample’s topography and refractive index. Despite the successful performance of DHM systems, their applicability to in-situ clinical research has been partially hampered by the need for a standard phase reconstruction algorithm that provides quantitative phase distributions without any phase distortion. This invited talk overviews the current advances in computational DHM reconstruction approaches from semi-heuristic to learning-based approaches.
The development of microscopic technologies has enhanced the translational research between scientists, engineers, biologists, and biomedical researchers, enabling the visualization and evaluation of complex biological systems. Advances in imaging systems are essential to further develop our understanding of cellular mechanisms and apply them to new diagnostic methods and disease treatment. Over the last decade, machine learning has been heavily used in microscopy image analysis, from the classification of cells to the reconstruction of real-time images, empowering the toolbox of automated microscopy. In this invited contribution, we discuss the use of generative adversarial networks in quantitative phase imaging and super-resolution microscopy.
KEYWORDS: Education and training, Data modeling, Holography, Holograms, Image restoration, Digital holography, Deep learning, Computer simulations, Microscopy, Environmental monitoring
Digital Lensless Holographic Microscopy (DLHM) is a phase imaging modality that omits the use of lenses or other bulky hardware to recover information from microscopic objects. Deep learning models have been recently used to substitute traditional DLHM reconstruction algorithms and classify samples from the reconstructed amplitude and phase images. In this work, we have investigated using these models to classify diatom samples, circumventing the whole reconstruction process altogether. We have validated our approach using a simulated DLHM dataset by comparing the performance of three typical image-processing learning-based models: AlexNet, VGG16, and ResNet-18.
Genetic algorithms enable optical system design with off-the-shelf optics. We propose a genetic algorithm that uses five-mutation operators for a scan lens design from scratch. This work demonstrates designs of multiple specifications and an as-built demonstrator.
KEYWORDS: Confocal microscopy, Signal to noise ratio, Data modeling, Point spread functions, Education and training, Microscopes, Performance modeling, Spatial resolution, Light sources and illumination, Super resolution
The advantages of confocal microscopy over widefield microscopy are their ability to produce optically sectioned images and their ability to produce multi-color imaging in which different organelles within the biological specimen are stained using multiple dyes, enabling colocalization studies. These features make confocal microscopy a widely used tool to provide valuable morphological and functional information within cells and tissues. One of the major drawbacks of confocal microscopy is its limited spatial resolution. Here, we train two generative adversarial networks using paired and unpaired data of low- and high-resolution images to improve the spatial resolution in confocal microscopy.
In this contribution, we present a study of the spatial resolution and quantitative phase imaging (QPI) performance of digital lensless holographic microscopy (DLHM) when holographic optical elements (HOEs) are employed as the illumination source of the system. The HOE employed in this study is a transmission hologram of a 3µm pinhole illumination system. We have quantified the imaging performance of the studied DLHM implementation by using a tailormade phase USAF test target. Results were then used to validate the dependence of the spatial resolution and phase sensitivity on the numerical aperture of the holographic point source.
This contribution presents a joint phase compensation and autofocusing method for telecentric off-axis Digital Holographic Microscopy (DHM). Current challenges of off-axis DHM systems applied to in-vivo imaging are the automatic reconstruction of phase images without phase distortions while the specimens under research move within the volume, generating out-of-focus holograms. Although different proposals to tackle these challenges individually have been reported for static samples, in this proposal, both issues are solved concurrently with no additional user intervention. As a result, in-focus compensated phase images of the out-of-focus studied samples are obtained. The proposal has been validated using simulated data.
Digital holographic microscopy (DHM), which provides quantitative phase imaging (QPI), has been widely applied in material and biological applications. The performance of DHM technologies relies heavily on computational reconstruction methods to provide accurate phase measurements. For example, non-telecentric DHM systems should compensate for the spherical wavefront associated with a non-telecentric configuration. The size of the ±1 diffraction orders in the hologram spectrum depends inversely on the radius of the curvature of the spherical wavefront introduced by the non-telecentric DHM system. Therefore, one can estimate the radius of curvature of the spherical wavefront by analyzing the hologram spectrum. Here, we outline the steps for the automatic reconstruction of phase images without distortions and with minimum user input from a hologram recorded in a non-telecentric DHM system. The proposed reconstruction approach can be divided into six main steps. The first step automatically selects the +1 diffraction order in the hologram spectrum. Secondly, the spherical wavefront parameters and the interference angle are estimated by analyzing the size and position of the selected +1 order. The third and fourth steps are the spatial filtering of the +1 order and the compensation of the interference angle, respectively. The next step involves the estimation of the center of the spherical wavefront. Finally, there is a fine-tuning step to optimize the estimated parameters and provide a phase image with minimum phase distortions. We have identified the relevant metrics in each step, compared multiple approaches, and selected the one with the higher performance for all our experimental holograms.
Polarization-sensitive microscopy has been a hot topic in biology over the last decades since many biological samples, including cancer cells and tendons, are intrinsically birefringent. Among its different applications, polarization-sensitive microscopy has enabled the detection and classification of diseases in biological samples by enabling the analysis of their polarimetric properties. In this contribution, we present an apparatus and method to estimate the Mueller matrix of microscopic samples using intensity-based images recorded from a bright-field microscope. We have validated the proposed technique using standard linear polarizers. The hallmark characteristics of our technique are its simplicity and efficiency, being suitable for analyzing biological and other birefringent samples.
One of the most common tools to recover the phase information of unstained microscopic translucent samples is Digital Holographic Microscopy (DHM). This imaging technique has a broad number of applications in biology and biomedicine. Nonetheless, to reconstruct an aberration-free phase image using DHM, a computationally demanding numerical process must be precisely executed. In this contribution, we present a generative adversarial network to fully compensate and reconstruct DHM holograms without the need for any computational process directly from the recorded hologram.
KEYWORDS: Holograms, Digital holography, Holographic interferometry, Video, 3D image reconstruction, Video acceleration, Cameras, Diffraction, Digital recording, Computer architecture
The quantification of the deformations presented by mechanical parts is a useful tool for several applications in engineering; regularly this quantification is performed a posteriori. In this work, a digital holographic interferometer for measuring micro-deformation at video rate is presented. The interferometer is developed with the use of the parallel paradigm of CUDA™ (Compute Unified Device Architecture). A commercial Graphics Processor Unit (GPU) is used to accelerate phase processing from the recorded holograms. The proposed method can process record holograms of 1024x1024 pixels in 48 milliseconds. At the best performance of the method, it processes 21 frames per second (FPS). This benchmark surpasses 133-times the best performance of the method on a regular CPU.
The numerical reconstruction of digitally recorded holograms has constituted the bottle neck for real-time digital
holography. The reconstruction process can be understood as the diffraction that undergoes a wavefront as it illuminates
the digitally recorded hologram. As this process is done numerically, the reconstruction of a M × N pixels hologram into
an image of similar dimensions is an operation with a Ο (M × N)2 complexity. The diffraction process can be represented
by a Fresnel transform or a scalable convolution of the recorded hologram. In these representations the numerical
reconstruction has a complexity of Ο (M × log N)2, still quite demanding computationally if the holograms are of 2048 × 2048 pixels. In this work, the power provided by a Graphics Processing Unit (GPU) is used to accelerate the numerical
reconstruction of digitally recorded holograms. The methodology is supported on the parallelization of typical Fresnel
transform and scalable reconstruction algorithms. On reconstructing holograms of 2048 × 2048 pixels, the reconstruction
is speeded up 20 times for the former method and 11 times for the scalable convolution. For holograms of 1024 × 1024,
the accelerated reconstruction methods allow for real-time digital holography.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.